Field of the Invention
The present invention relates to an image blur correction device, an imaging apparatus, a control method of the imaging apparatus, and a non-transitory storage medium.
Description of the Related Art
As one photography method of a camera, there is panning. Panning is a photography scheme of setting a shutter speed to be slow and taking a picture while causing a camera to follow a motion of an object moving in, for example, a horizontal direction to obtain a feeling of lively motion of the object. In panning, when blur occurs between a motion of an object and a motion of a camera, object blur occurs. As a shutter speed is slower, a panning amount in a background increases. Thus, a feeling of lively motion of an object can be obtained. However, image blur or object blur easily occurs.
Accordingly, two photography assist functions are considered to simply realize panning. The first assist function is a function of detecting blur between a motion of an object and a motion of a camera and correcting object blur according to the blur with an optical correction system (hereinafter referred to as a first assist function). The second assist function is a function of automatically setting a shutter speed in a camera (hereinafter referred to as a second assist function). A photography mode that has the two assist functions is called a panning assist mode (hereinafter referred to as panning assist).
Japanese Patent Laid-Open No. 2006-317848 discloses an imaging apparatus that detects a difference between a speed of an object and a blur speed of a camera and corrects image blur equivalent to the difference with regard to the first assist function using an image blur correction function.
The imaging apparatus disclosed in Japanese Patent Laid-Open No. 2006-317848 detects a vector equivalent to an object (hereinafter referred to as an object vector) among vectors detected with a motion vector using the motion vector and an angular velocity and detects a difference between a speed of the object and a blur speed of the camera. The difference between the speed of the object and the blur speed of the camera corresponds to blur between a motion of the detected object and a motion of the camera. Known examples of a method of detecting a motion vector are a block matching method and a correlation method based on a correlation arithmetic operation. In the bock matching method, an input image signal is divided into a plurality of block regions with an appropriate size and a motion vector is detected for each block. First, an image signal is divided into block regions, differences from a previous frame are calculated in units of blocks, and a block of a frame before a sum of absolute values of the calculated differences is the minimum is retrieved. Then, a relative deviation between screens of a certain block of a current frame and the retrieved block of the frame before the sum of the absolute values of the differences is the minimum indicates a motion vector of the block.
However, for example, in vector detection of a pattern with no feature point, such as a pure white wall, a motion vector is erroneously detected in some cases. When a motion vector is erroneously detected, a driving amount of an optical correction system may not be correctly calculated. Therefore, there is a concern of object blur correction being controlled excessively.